About / Risk Management, Research & Data Science

Evidence before certainty.

I am a risk and data analytics professional with 5+ years of experience in financial-risk workflows and a background in economics, statistics, and data science. My work connects reliable data, transparent models, disciplined controls, and clear communication so decisions remain useful under uncertainty.

Background

Experience across risk and delivery.

Roles across financial risk analytics and earlier online trading firms show how I move from technical detail to reliable execution, commercial awareness, and clear stakeholder communication.

2020 — Present

Financial risk analytics

Hedge-Tech | Fin-Tech & Risk Management

Analyze large financial datasets; review and improve automated risk-analysis scripts; validate inputs, outputs, assumptions, and reporting logic; support financial-instrument modelling, ALM/liquidity and uncertainty workflows; investigate data and process issues; and deliver stable, documented outputs for major financial institutions.

The role also requires coordination with client-side technology teams, regulation-related reporting support, root-cause analysis, issue escalation, and clear recommendations for both technical and business stakeholders.

2017 — 2019

Project & stakeholder management

GTP Solutions | Online trading firm

Coordinated projects in an online trading environment across sales and marketing activity, team workflows, stakeholder communication, and client and employee relationships. Followed delivery across multiple priorities and kept commercial and operational work moving clearly between stakeholders.

2015

Client-facing sales

DMD Online | Online trading firm

Handled client-facing sales and service in an online trading environment, explaining offerings, identifying customer needs, and maintaining clear follow-up throughout the sales process.

MA

Statistics and Data Science · In progress

Statistical modelling, machine learning, optimization, stochastic processes, simulation, predictive analytics, deep-learning methods (MLPs, CNNs, and RNNs) with PyTorch and TensorFlow, and model evaluation.

BA

Economics and Statistics · Completed

Economics, probability, inference, quantitative reasoning, financial markets, and data-informed decision-making.

Practice

Risk, research and data science.

Risk management defines exposures, controls, limits, and accountability. Statistics measures evidence, uncertainty, and predictive performance. Research connects both to stronger questions, transparent methods, and practical decisions.

Risk practice

Risk management

Financial-risk analysis, controls, modelling, validation, and decision-ready reporting.

  • Liquidity & ALM: ALM and maturity ladders, LCR, NSFR, ASF/RSF, cash-flow gaps, funding stability, limits, and liquidity scenarios.
  • Credit risk: PD/LGD/EAD, expected loss, scorecards, limits, vintage and cohort monitoring, and early-warning signals.
  • Market risk: Instruments, sensitivities, VaR and Expected Shortfall foundations, limits, stress scenarios, and portfolio interpretation.
  • Enterprise & operational: Risk appetite, KRIs, controls, incidents, IT/data risk, root-cause analysis, and remediation.
  • Model & reporting risk: Assumptions, benchmarks, backtesting, monitoring, data lineage, reproducibility, and escalation.
Research & modelling

Statistics & data science

Applied statistics, research, predictive modelling, validation, and reproducible analysis.

  • Statistical methods: Probability, inference, hypothesis testing, regression, stochastic processes, simulation, and uncertainty.
  • Predictive modelling: Trees, Random Forest, XGBoost, multilayer perceptrons (MLPs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), segmentation, and calibrated probabilities.
  • Validation: Leakage-safe cross-validation, benchmarks, metrics, calibration, diagnostics, feature importance, and stability.
  • Research & experimentation: A/B testing, sample-size reasoning, causal foundations, cohorts, funnels, and root-cause analysis.
  • Data workflows: EDA, cleaning, validation, automated checks, reproducible code, dashboards, and technical reporting.
The shared ground

Where risk and statistics become one practice

Uncertainty

Scenarios

Sensitivity, Monte Carlo, baseline/adverse/severe scenarios, reverse stress, tail-risk reasoning, and explicit assumptions.

Models

Evidence with controls

Fit and predictive quality are paired with validation, monitoring, limitations, governance, human checkpoints, and defensible use.

Decisions

Signals people can act on

Complex results become limits, alerts, risk queues, operating thresholds, clear trade-offs, and concise stakeholder recommendations.

Risk method set

Controls around the calculation.

Core methods for testing models, understanding exposures, monitoring limitations, and producing reliable risk decisions.

Core practice

Model validation, backtesting & monitoring

Test conceptual soundness, implementation, outcomes, stability, and limitations; compare observed performance with expectations and benchmarks; and document required action.

  • Backtesting
  • Benchmarking
  • Outcome analysis
  • Calibration
  • Drift & stability
  • Sensitivity analysis
  • Stress testing
  • Limitations
  • Issue tracking
Risk modelling

Financial exposures & scenarios

Model cash flows, funding, credit and market exposures across ordinary, adverse, and reverse stress conditions.

  • ALM / maturity ladder
  • LCR
  • NSFR
  • ASF / RSF
  • PD · LGD · EAD
  • Expected loss
  • VaR / Expected Shortfall
  • Scenario analysis
  • Monte Carlo
  • Uncertainty modelling
Controls & reporting

Governance around the output

Keep risk work accountable through limits, controls, data quality, reproducibility, documentation, escalation, and tracked remediation.

  • KRIs
  • Limits & thresholds
  • Data lineage
  • Reproducibility
  • Reconciliations
  • Automated reporting
  • Root cause & remediation
  • Regulatory documentation
Working toolkit

From raw data to a defensible answer.

The technical stack supports the same operating principle throughout: validate first, automate repeatable work, document assumptions, and communicate the decision rather than only the model.

Programming & analysis

Python · SQL · R

pandas, NumPy, scikit-learn, statsmodels, PyTorch, TensorFlow, Matplotlib, seaborn, data extraction and cleaning, statistical modelling, visualization, script review, and workflow automation.

Reporting & productivity

Excel · BI · documentation

Excel, Power BI, Tableau, dashboards, analytical presentations, Git, Markdown, Obsidian, reproducible notes, methodology documentation, and stakeholder-ready reporting.

AI-assisted workflows

Automation with oversight

Retrieval-augmented generation (RAG), LLM-assisted research, prompt engineering and testing, code and output review, documentation automation, workflow design, responsible AI use, and explicit human judgment at decision points.

QA & reproducibility

Testing mindset

Manual and automated checks, test-case thinking, input/output validation, regression checks, reproducibility review, issue documentation, data-quality investigation, and controlled change.

Business delivery

Clarity across functions

Client service, stakeholder communication, technical storytelling, presentations, team and project coordination, sales/marketing awareness, fast ownership, and action-oriented writing.

Leadership & context

Multilingual, cross-cultural work

Arabic, English, and Hebrew; public speaking, negotiation, mediation, research, diplomacy, cross-cultural dialogue, teamwork, and structured conflict resolution.

See the methods in context.

Return to the project portfolio for applied examples of predictive modelling, calibrated risk ranking, threshold decisions, cross-validation, and decision-focused reporting.

Back to main page